Political Party Positioning in Referendums. Introducing a theoretical framework

Paper Proposal S17 – Democratic Innovations 12 Referendums and Policies: Explaining Choices, Challenges, and Impact Political Party Positioning in Referendums. Introducing a theoretical framework In the literature on direct democracy, party cues are considered to have a significant influence on the voting decision of the electorate. But how do parties come to their decisions in a direct-democratic vote in the first place? What are the determinants of party positioning in direct democracy? What role does the institutional framework play? Referring to the classic literature about party goals, the paper presents a framework on the determinants and processes of parties’ decision-making in a direct-democratic event. Two different models can be identified: the policy and the responsiveness model. 1) In the policy model, parties take their decisions very early, usually in the parliamentary arena. The majority of direct-democratic instruments in use are top-down and also bottom-up instruments usually include a parliamentary step. In this model, voting issues touch upon the traditional domains of party politics. Due to this high degree of issue partyness, the path dependency of former policy positions determines party positioning. Alternatively, also or non-political, technical policy consideration may play a role. Path dependency of previous decisions and technical/non-political constraints also prohibit a change of the first positioning decision. This position stability applies to all party units. Responsiveness as a motive for positioning does not apply because public opinion on the voting issue and issue saliency are still undetermined at this early stage. 2) In the responsiveness model in contrast, the parties take their positions early as well, but early party positions are unstable. The issue partyness is lower or absent, allowing for party responsiveness to media pressure, public opinion, or campaign events. In this model, parties do not switch their decisions neither, but hollow out, soften up, or moderate them later in the campaign. The party signalizes to the followers and the public, that the formal decision is only relevant on paper. In the second part, the paper offers a first application of the framework to Swiss Direct Democracy. An original dataset of 62 votes from 2013 to 2019, five political parties with 310 cases of party positioning is compiled. The measurement of the different variables uses the Swissvotes-Dataset, the Party-Manifesto-Dataset, and the Foeg-Monitor. The mainly descriptive results confirm the relevance of the two models. Swiss parties decide in the parliamentary arena way before the beginning of the campaigns. They do not change their positions until the voting day with only little deviations from cantonal party units. But occasionally Swiss parties moderate their positions in the course of the campaigns. Path-dependence of former policy positions and non-political motives each seem to account for almost half of the votes whereas responsiveness to campaign events occurs only seldomly. In conclusion, the paper discusses the applicability of the framework to other national experiences of direct democracy.